The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines
By Luiz André Barroso, Urs Hölzle. 2009
1,992 citations recorded by the source corpus.
As computation continues to move into the cloud, the computing platform of interest no longer resembles a pizza box or a refrigerator, but a warehouse full of computers. These new large datacenters are quite different from traditional hosting facilities of earlier times and cannot be viewed simply as a collection of co-located servers. Large portions of the hardware and software resources in these facilities must work in concert to efficiently deliver good levels of Internet service performance, something that can only be achieved by a holistic approach to their design and deployment. In other words, we must treat the datacenter itself as one massive warehouse-scale computer (WSC). We describe the architecture of WSCs, the main factors influencing their design, operation, and cost structure, and the characteristics of their software base. We hope it will be useful to architects and programmers of today's WSCs, as well as those of future many-core platforms which may one day implement the equivalent of today's WSCs on a single board. Table of Contents: Introduction / Workloads and Software Infrastructure / Hardware Building Blocks / Datacenter Basics / Energy and Power Efficiency / Modeling Costs / Dealing with Failures and Repairs / Closing Remarks
Subfields: Computer Networks and Communications, Information Systems, ML Systems & Hardware
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Networks and Communications
- Christos Kozyrakis
- Ion Stoica
- Christina Delimitrou
- Onur Mutlu
- Young Geun Kim
- Carole-Jean Wu
- Liqun Cheng
- Rama K. Govindaraju
In Information Systems
- Christos Kozyrakis
- Jimmy J. Lin
- Christina Delimitrou
- Ion Stoica
- Liqun Cheng
- David Lo
- Rama K. Govindaraju
- Mu Li
In ML Systems & Hardware
- Jimmy J. Lin
- Mu Li
- Christos Kozyrakis
- Alex Smola
- Carole-Jean Wu
- Young Geun Kim
- Ion Stoica
- Liqun Cheng
Citations in the corpus
- Chasing Carbon: The Elusive Environmental Footprint of Computing (2020, background): “These characteristics must consider the lifetime of systems; for example, data centers typically maintain server-class CPUs for three to four years [13].”
- PARTIES: QoS-Aware Resource Partitioning for Multiple Interactive Services (2019, background): “Cloud computing has become ubiquitous by ofering resource lexibility and cost eiciency [9, 11, 31].”
- Data Center Energy Consumption Modeling: A Survey (2016, methodology): “An approximate distribution of peak power usage by components of a warehouse scale computer deployed at Google in 2007 [102].”
- Characterizing cloud computing hardware reliability (2010, background): “We hope that the results presented in this paper will serve as motivation to foster further research in this area.”
- Megastore: Providing Scalable, Highly Available Storage for Interactive Services (2011, background): “Economically constructed sites risk some level of facility-wide outages [25] and are vulnerable to regional disasters.”
- FIRM: An Intelligent Fine-Grained Resource Management Framework for SLO-Oriented Microservices (2020, neutral): “, latency spikes [4,11,22,30,35,44,53,69,98,99].”
- Optimizing Data-Center TCO with Scale-Out Processors (2012, background): “A scale-out data center powering cloud services can house tens of thousands of servers that are necessary for high scalability, availability, and resilience.(1) The massive scale of such data centers requires an enormous capital outlay for infrastructure and hardware, often exceeding $100 million pe”
- Page placement in hybrid memory systems (2011, background): “As a result, the amount of energy consumed by the memory is approaching (and sometimes surpassing) that consumed by the processors in many servers [2, 21].”
- Bubble-up: Increasing utilization in modern warehouse scale computers via sensible co-locations (2011, background): “INTRODUCTION “Warehouse scale computers” (WSCs) [12,22] house large scale web applications and cloud services.”
- Quasar: resource-efficient and QoS-aware cluster management (2014, methodology): “We present Quasar, a cluster management system that increases resource utilization while providing consistently high application performance.”
- TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddings (2023, background): “Computer architects improved PUE by helping advance the state-of-the-art of warehouse scale computers (WSCs) [2, 3, 16, 30, 44, 62].”
- Heracles: Improving resource efficiency at scale (2015, methodology): “Several studies have established that the average server utilization in most datacenters is low, ranging between 10% and 50% [14, 74, 66, 7, 19, 13].”
- A survey and taxonomy on energy efficient resource allocation techniques for cloud computing systems (2016, background): “This is due to the fact that the energy required by the datacenters for its operation, illumination, power supply, and cooling, contribute significantly to the total operational costs [13].”
- Silicon CMOS-integrated nano-photonics for computer and data communications beyond 100G (2012, background): “To meet these demands, the concept of warehouse-scale datacenters (WSCs) is being developed [2] with large numbers of commodity servers interconnected by massive numbers of high-bandwidth optical links [3].”
- Predicting the Performance of Virtual Machine Migration (2010, background): “5) MapReduce Tasks: There is currently wide-spread interests in the MapReduce paradigm for large-scale data processing and analysis [19], [ 20 ].”
- Inside the Social Network's (Datacenter) Network (2015, background): “The precise degree of oversubscription varies, but there is general agreement amongst operators that full connectivity is rarely worthwhile [11].”